🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the NVIDIA DGX Spark run Mixtral 8x7B?
Yes — here's how fast and up to what context, with real measured numbers.
Mixtral 8x7B on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 62k | 2.1-2.5 t/s | ~0.1-0.1 (slow) |
| Q8 | yes | 204k | 4.3-5.1 t/s | ~0.2-0.2 (slow) |
| Q4 | yes | 267k | 7.6-9.1 t/s | ~0.3-0.3 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the NVIDIA DGX Spark run Mixtral 8x7B?▶
Yes. At Q4 it generates ~7.6-9.1 tok/s and fits up to 267k context; beyond that it spills to SSD swap and slows to ~0.3-0.3 tok/s.
How much context fits?▶
At Q4, up to about 267k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.